Spread spectrum is a method of modulation that needs a transmission bandwidth usually substantially higher than data rate. In a direct sequence spread spectrum communications system, the transmitter modulates a data signal with a pseudo random chip sequence to generate spread spectrum signal. Usually the chip rate of the pseudo random sequence is much higher than the data rate and therefore the direct sequence spread spectrum communications system will take substantially wider bandwidth than the bandwidth needed by data signal itself. The spread spectrum signal is then transmitted over a communications media as a radio wave to a receiver. The receiver despreads the spread spectrum signal to recover the information contained in the received spread spectrum signal.
Having many advantages over other communications systems, direct sequence spread spectrum communications system is one of the major communications systems widely used in today's society. However, spread spectrum communications system has some disadvantages. One of the major disadvantages is low spectrum efficiency. Nowadays more and more applications require higher and higher data rate but the available bandwidth is both very expensive and limited. The low spectrum efficiency inherited in a spread spectrum communications system will greatly restrict its opportunity to be used in many high data rate applications.
Multi-channel direct sequence spread spectrum communications system is one of the attempts to increase the data rate within a given bandwidth and therefore to improve the spectrum efficiency of direct sequence spread spectrum communications system. However, in a regular multi-channel direct sequence spread spectrum communications system, the more channels, the more different signal levels. In order to have enough resolution, more linear power amplifier and higher precision A/D converter will be needed. Both of these factors will make cost greatly increased especially when the working clock is very high.
U.S. Pat. No. 6,324,209 to Don Li et al, herein incorporated by reference, disclosed a method of applying multi-channel technology in direct sequence spread spectrum communications system. Different from regular multi-channel direct sequence spread spectrum communication system in which each sub-channel transmits information simultaneously, U.S. Pat. No. 6,324,209 separated spreading codes into several subgroups and at any time, transmitted only one spreading code from each subgroup. In this way, it limited the possible number of signal levels and therefore the requirement on both A/D and linear power amplifier was under control. However, the data rate was still not high enough. In some case, it can not meet the requirements such as reasonable cost and limited transmission bandwidth set up by more and more high data rate applications.
Therefore, there is a need to further increase the spectrum efficiency of a multi-channel direct sequence spread spectrum communications system without increasing the requirement on linear power amplifier and A/D converter.
In a regular communications system, the information is represented and transmitted directly by each transmittable signal. Let S be the set of all these transmittable signals. The power set of the set S is a set denoted by 2S, whose elements are the subsets of S. In other words, 2S={X|X⊂S}. If there are L components in S, then there will be 2L elements in the power set 2S.
Suppose in a regular communications system, each element of S can carry 1 bit information. By transmitting all of the elements of S, one can send L bits information.
One can also send L bits information by alternative. Due to the fact there are 2L elements in the power set 2S, one can correspond each element of the power set 2S to a different L bits binary number. Since there are CLi elements in 2S with each element consisting of i components from set S, the element of 2S on average consists of
            (                        ∑                      i            =            0                    L                ⁢                  i          ·                      C            L            i                              )        /          2      L        =      L    /    2  components from set S. In this alternative way, instead of transmitting L signals all the time, one can, on average, transmit only half of L signals.
Furthermore, instead of using all the elements of a power set to transmit information, one can use a particular subset of the power set to transmit information.